¡En Vivo Contigo!

$250-750 USD

Posted 9 minutes ago

•

Ends in 6 days

$250-750 USD

Paid on delivery

I want a complete n8n workflow that funnels every website-chat message through two AI agents—Anthropic Claude and Google Gemini—so visitors get instant, accurate answers about their own account, order, or booking. Here’s how I picture the flow: a webhook catches the chat event, n8n verifies the customer’s identity, pulls the relevant account or order details from our database/CRM, then feeds that context to Claude. If Claude’s confidence score drops below an agreed threshold or a policy rule is triggered, Gemini steps in as either a fallback or a second opinion. The final answer is returned to the chat widget in real time, all within a few seconds. Key deliverables • Production-ready n8n workflow file (.json) with clear node naming and comments • Secure connectors to our database/CRM (REST or SQL, whichever you choose) that surface only the fields required for support answers • Claude and Gemini prompt templates that inject the retrieved customer data and enforce tone/formatting guidelines • Error-handling, retry, and logging nodes so failed calls are flagged and stored for review • Environment variable setup and a quick-start README so I can import the flow, drop in my API keys, and go live • Demo recording or live walkthrough proving the agent answers typical chat questions—“Where’s my order?”, “What’s the status of booking #1234?”, “Can I change my delivery address?”—with no manual intervention Acceptance criteria 1. A visitor receives a correct, personalised reply in under 5 seconds. 2. Sensitive data never appears in logs or model prompts unless explicitly required for the answer. 3. Switching model order (Gemini first, Claude fallback) is a one-line change in the workflow. If you’ve built multi-model automations in n8n before—or have clever ideas on prompt engineering, vector stores, or function calling to sharpen responses—I’m all ears. Let’s ship this fast.

Project ID: 40821682

About the project

146 proposals

Open for bidding

Remote project

Active 35 secs ago

Place your bid

Bid amount Email address

Benefits of bidding on Freelancer

Set your budget and timeframe

It's free to sign up and bid on jobs

146 freelancers are bidding on average $389 USD for this job

User Avatar
Flag of UNITED STATES

Hi, I am Elias from the sunny state of Florida. I understand that your goal is to create a complete n8n workflow that effectively channels website chat messages through two AI agents—Anthrop. This integration is crucial for enhancing user interaction while ensuring the process is efficient and reliable. To approach this, I would first analyze your existing chat infrastructure and how n8n can be utilized to facilitate the communication between your website and the AI agents. I'll focus on creating a robust workflow that accurately captures messages, processes them through the AI agents, and returns responses seamlessly. This involves setting up appropriate API integrations and ensuring that the data flow adheres to your requirements. A few questions to clarify the scope: Q1 – What specific functionalities do you want the AI agents to perform with the chat messages? Q2 – Are there any existing systems or platforms that we need to integrate with n8n? Q3 – What is your preferred method for handling user data and privacy considerations? I look forward to discussing this project further and helping you achieve your vision. Looking forward to hearing from you.

User Avatar
Flag of PAKISTAN

Hello, Building secure AI-powered automations and CRM/API integrations has been central to my work for over 15 years, making me well suited to develop RespondeYA! as a reliable n8n support workflow. My experience with Python, AI automation, REST APIs, ERP/CRM systems, SaaS platforms, MySQL, PostgreSQL, and cloud deployment aligns directly with your webhook, customer-data retrieval, model orchestration, and production-readiness requirements. I can create the importable n8n JSON workflow with clearly named nodes, Claude and Gemini integrations, prompt templates, configurable model priority, identity-aware account/order lookups, retries, error handling, and review logging. My API Development and third-party integration experience supports secure data exchange while keeping retrieved fields focused on what each answer needs. I will also provide environment-variable guidance, a quick-start README, and a walkthrough covering order, booking, and delivery-address scenarios. I focus on clean, maintainable solutions that are practical to deploy and support long term. Choose me to turn your multi-model customer-support concept into a polished automation ready for real-world use. Thanks!

User Avatar
Flag of PAKISTAN

Hello, I can build the complete n8n workflow for RespondeYA!, connecting the chat webhook, identity verification, CRM or database lookup, Claude, and Gemini fallback logic. I’ll structure the workflow with secure field selection, prompt templates, retries, error handling, redacted logging, environment variables, and clear documentation. I’ll also make model priority configurable through a single workflow setting and validate the under-five-second response target using realistic order, booking, and delivery-address scenarios. A useful improvement would be adding explicit tool/function boundaries so each agent can access only the customer data required for its current intent. I can provide the import-ready JSON, README, and walkthrough using your preferred CRM or database connector. Regards Muhammad

User Avatar
Flag of INDIA

Hi, I will build your n8n workflow so every chat message is verified, enriched with the customer’s exact account/order/booking data, then answered by Claude with Gemini as fallback or second opinion when confidence is low or policy triggers. I’ll include secure connectors, prompt templates, retries, logging, clean node naming, env setup, and a ready-to-import JSON plus README. Start the chat so we can discuss the flow in detail. Regards, Sai Bhaskar

User Avatar
Flag of PAKISTAN

Greetings, I see that you’re looking to create an n8n workflow that channels chat messages through two AI agents—Claude and Gemini—to give customers quick and accurate responses about their accounts and orders. My approach would involve setting up a webhook to capture chat events, verifying user identity, and pulling in relevant data from your CRM. This context would then be sent to Claude, with a fallback to Gemini if needed, ensuring a seamless user experience. With my experience in PHP, API development, and n8n integrations, I can build a robust workflow that meets your criteria. I will implement secure database connections, manage error handling, and create clear documentation for easy setup. The ability to easily switch model orders will also be included, making adjustments straightforward. Best regards, Saba Ehsan

User Avatar
Flag of UNITED STATES

Hi, the critical part is ensuring each visitor can retrieve only their own records. I’d verify the website’s authenticated session server-side and enforce account ownership before passing minimal context to either model. In n8n, I’d build a clear webhook -> identity check -> scoped lookup -> primary model -> validation/fallback -> response flow. Model order would use one configuration value, making Claude/Gemini switching straightforward. Fallback decisions would combine defined policy checks, missing evidence and output validation rather than rely solely on a model’s confidence score. Order and booking statuses would come from your systems; address changes would require explicit permission and confirmation. The under-five-second target needs early benchmarking, especially when two models run sequentially. I’d measure typical and fallback paths separately, use timeouts and avoid unnecessary lookups. Delivery would include the documented JSON, secure connectors, both prompts, redacted logs, retry/error handling, environment setup, README and demo. Tests would cover cross-account requests, duplicate events, unavailable services and common support questions. I’d estimate 5–8 working days after reviewing your integrations. Which chat widget, CRM/database and login system do you use, and should address changes be executed or only explained?

User Avatar
Flag of BULGARIA

Hello, I'm excited to help you create a smooth n8n workflow that effectively routes chat messages through both Claude and Gemini for instant support. It sounds like you want a system that not only verifies customer identity but also ensures accurate, timely responses while maintaining data security. With 7+ years of experience in API development and chatbot integration, I can build a solution that meets your needs. My approach would involve setting up the webhook to capture chat events, structuring the workflow for easy model switching, and ensuring that sensitive data remains protected. I’ll also make sure to include comprehensive error handling and clear documentation for your team. One question I have is about the specific tone or style you want the AI responses to have. This will help me fine-tune the prompt templates for Claude and Gemini. You can check out my work at https://pixoriq.com. Best regards, Ivan Mandinski

User Avatar
Flag of PAKISTAN

Hi there! I’ll build a production-ready n8n workflow that captures every website chat message, verifies the customer, pulls only the minimum account/order/booking fields needed, and routes the context through two AI agents (Claude and Gemini) with confidence/policy-based fallback. You’ll be able to switch the model order with a one-line workflow change. My approach: - Deliver an importable n8n .json with clear node names/comments - Add secure identity verification plus REST/SQL data connectors for support-safe fields only - Include Claude and Gemini prompt templates that enforce your tone/format rules - Implement retries, error-handling, and logging that redacts sensitive data For proof, I’ll include a short live walkthrough (or demo recording) covering “Where’s my order?”, “booking #1234”, and delivery address changes. Regards, Umer Hayat

User Avatar
Flag of INDONESIA

Hi there, Your vision for RespondeYA! is exactly the kind of multi-model automation I specialize in. I’ve previously designed n8n workflows integrating AI agents like Claude and Gemini, handling secure data retrieval from MySQL/REST APIs, and delivering instant, context-rich responses for support teams. Your requirements for real-time, accurate answers—with secure handling of sensitive data and robust error management—align perfectly with my expertise. Here’s my approach: I’ll build a modular n8n workflow where chat events trigger identity verification and precise data extraction from your database/CRM, limiting exposed fields to only those necessary for the AI’s response. Prompts for Claude and Gemini will inject this context, with tone and formatting strictly enforced. Confidence scores and policy checks will govern fallback logic, and switching model order will be as simple as toggling a variable. Every step will include detailed comments and clear node naming for easy maintenance. Error-handling, retry mechanisms, and secure logging will ensure reliability and compliance. Environment variables will keep credentials safe and deployment simple. I’ll provide a step-by-step README and a recorded demo showing personalized, instant answers to real support queries. If you’re interested, I can also advise on advanced enhancements like vector-based retrieval or function-calling for even sharper AI performance. Let’s create a seamless, intelligent chat experience that delights your users and scales with your needs. Best Regards

User Avatar
Flag of INDIA

The part that decides whether RespondeYA! works is the identity check plus minimal context retrieval before 'Claude’s confidence score' is evaluated, so I’ll design that gate before wiring the model fallback. I’ll build the n8n webhook flow with REST or SQL database/CRM nodes, env variables for API keys, redacted logging, retry/backoff, and error branches that store failures without exposing sensitive fields. I’ll add Claude and Google Gemini prompt templates with required account/order/booking fields only, a confidence/policy switch, and a one-line model-order setting so Gemini can run first if needed. You’ll receive the importable .json workflow, README, prompt templates, and a demo recording or live walkthrough showing under-5-second replies for the sample chat cases. Send the database/CRM API details, chat webhook format, and Claude/Gemini keys; I can start now. Rajavel n8n Automation Developer | Claude | Google Gemini | API Integrations

User Avatar
Flag of PAKISTAN

I will build a production-ready n8n workflow connecting Claude and Gemini with secure customer verification, database/CRM retrieval, and personalized real-time chat responses. I’ll implement confidence-based fallback, configurable model order, privacy-safe prompts, retries, error handling, and logging, then deliver the importable JSON, prompt templates, environment setup, README, and a working demo covering your acceptance scenarios. Best Regards, Muhammad

User Avatar
Flag of NORWAY

Before any prompt gets written I'd put the model-order switch and a PII-redaction node in place. Every Claude or Gemini call then runs through the same redaction step, and changing the primary model is a one-line edit. After that comes the chat webhook, customer verification and the DB/CRM lookup. Claude answers first, and Gemini takes over on low confidence or a policy hit. Every branch gets error handling and logging. The demo recording runs order status, booking #1234 and the address change against mock data. M1: webhook, verification, redaction and model switch, $211. M2: DB/CRM lookups, Claude and Gemini fallback, prompt templates, $215. M3: error handling, logging, README and demo, $214. The $640 holds if there is one chat widget webhook and one DB or CRM I can reach over REST or SQL. Customer, order and booking data must already be queryable by ID, because I won't design the schema. You'd supply the Claude and Gemini keys and an n8n instance to import into. If any of that differs, I re-quote before starting work. Which chat widget are you on, and what does its webhook payload look like? Is the data in a database or a CRM? Where should a low-confidence answer go if Gemini is unsure too? Does the 5 seconds include the widget's own delay?

User Avatar
Flag of GERMANY

Hi, With 15+ years of experience and 100+ projects, including work for the Max Planck Institute in Munich and US Department of Defense, I specialize in AI automation, API integrations, and production workflows. I can build your RespondeYA! n8n solution with secure multi-model AI orchestration. My approach: • Capture chat events through authenticated webhooks. • Verify customer identity and retrieve CRM/order data. • Use Claude for contextual responses, with Gemini fallback. • Implement confidence checks, retries, logging, and error handling. • Optimize response latency toward your 5-second target. • Deliver importable n8n JSON, prompts, environment setup, README, and demo recording. Best regards, Iosif Peterfi

User Avatar
Flag of GEORGIA

Hello — the n8n chat routing flow looks like the kind of system that needs to be reliable from the first message all the way through handoff. I understand the goal here is to send every website chat message through two AI agents, then keep the logic clean enough that it works smoothly with your WordPress/PHP/MySQL setup. The Anthropic part, the message routing, and the integration details matter more than just “making it work” once. I’ve built similar systems with n8n, API integrations, WordPress, PHP, and MySQL, and I’ve also worked on customer-support assistants and AI agent workflows that pull from business data and route requests intelligently. I’m comfortable with webhook-based flows, backend glue code, and making sure the workflow is maintainable after launch. If helpful, I can share relevant examples in chat. A couple things I’d want to confirm: Which chat system or widget is sending the messages today? What should the two agents do differently, and what decides the handoff between them? Happy to talk through the flow with you and map it out cleanly.

User Avatar
Flag of UNITED STATES

Hi, I can build your production-ready n8n support workflow using Claude, Gemini, and secure CRM integrations for $550 within 5 days. I have hands-on experience with multi-agent AI automation, REST APIs, databases, and webhook orchestration. I'll focus on accurate personalized answers, secure customer data, and fast responses. My approach uses n8n with Claude as the primary agent and Gemini as a configurable fallback. I'll connect your website chat, validate customer identity through your existing authentication system, retrieve only authorized account or order details, and generate context-aware replies without exposing unnecessary personal information. I'll implement configurable model routing, policy-based escalation, structured confidence checks, retries, error logs, and secure credential management. The workflow will support order and booking inquiries, with sensitive account changes requiring verified authorization. I'll test response latency against your five-second target and document any external API limitations. My fixed bid is $550 for 5 days, including importable n8n JSON, prompt templates, setup instructions, testing, and a demo walkthrough. I'm ready to begin after award and initial milestone funding.

User Avatar
Flag of PAKISTAN

Hi, I got that you are looking for a production-ready n8n workflow that routes chat messages through Claude and Gemini, verifies the customer, pulls the right account or order data, and returns a fast personalized reply with secure logging and fallback logic. This is what I can help you with, let's chat To handle your project, I will build the workflow in n8n using webhook triggers, secure HTTP/SQL connectors, and structured prompt templates for Claude and Gemini with confidence-based routing and policy checks. I’ll design the hardest part, the multi-model fallback with sub 5 second personalized responses, so Gemini can act as a second opinion or primary model with a one-line switch, while sensitive fields are filtered before prompts and logs. You’ll get reliable identity verification, targeted data retrieval, retry handling, and error flags without exposing unnecessary data. I’ll also test the exact questions you listed and refine the prompts before handoff. As final deliverables you will receive the complete .json workflow with clear node names and comments, secure database or CRM connections, Claude and Gemini prompt templates, error handling and logging nodes, environment variable instructions, a quick-start README, and a demo recording or live walkthrough showing the bot answering real chat scenarios without manual intervention. Let’s connect and I’ll map the flow cleanly. Regards, Imran

User Avatar
Flag of PAKISTAN

Hello, I’m Maryam Abbas, with 6 years of experience in WordPress, n8n, PHP, HTML, MySQL, AI Integration, and API Development. I’ve read your requirement carefully. I can build a production-ready n8n workflow that securely verifies users, pulls only the needed account/order data from your CRM or database, and routes each chat message through Claude with Gemini as fallback/second opinion based on confidence or policy rules. I’ll include clean node naming, prompt templates, retries, logging, error handling, environment setup, and a quick-start README so you can import, configure API keys, and launch smoothly. I’ve accumulated over 6 years of experience working with various technologies. It has been a rewarding journey since I started 6 years ago. I invite you to review my profile, focusing on Skills, Portfolio, and Client Testimonials. Why Choose Me? I’ve delivered numerous successful projects, and for the past 4 years I’ve consistently received positive feedback with no negative reviews. Please start the chat so we can discuss it further. Best regards, Maryam Abbas

User Avatar
Flag of PAKISTAN

Hello, I can build your production-ready n8n workflow for RespondeYA!, routing each chat request through secure identity verification, CRM/database retrieval, Claude, and Gemini fallback or second-opinion logic. The implementation will include clearly named and commented nodes, minimal-field data access, privacy-safe prompts and logging, retries, failure alerts, environment-variable configuration, and a practical README. I’ll also structure model selection so switching the primary and fallback order is straightforward. Performance will be designed around your under-five-second target, with attention to webhook response handling and efficient API/database calls. I’ll validate the workflow against order tracking, booking status, and delivery-address-change scenarios, then provide a walkthrough demonstrating the complete flow and escalation behavior. Sensitive values will be excluded from logs and passed to models only when necessary. Before implementation, I’d be glad to confirm the integration details and acceptance expectations. Could you please share which chat widget, CRM/database, and customer-authentication method the workflow must integrate with? Best regards, Talha

User Avatar
Flag of CANADA

I’ve built n8n workflows that handle multi-step API calls combining chatbots with real-time data fetching, so I understand the importance of speed and accuracy in customer conversations. Your plan to verify identity, fetch specific account/order info, then send it to Claude with Gemini as fallback makes sense. I’d set up secure, least-privilege database connectors that expose only needed fields and ensure no sensitive info leaks into logs or prompts unless necessary. For prompt design, I suggest templating to tightly control tone and info density, plus dynamic threshold checks on Claude’s confidence score for seamless failover to Gemini. A single config node to reorder agents is easy to add so you can toggle which model leads. Error handling with retries and logging failed events will keep things robust. Are your database/CRM APIs mostly REST or SQL? That’ll help me choose the best connector upfront. Also, do you want partial answers when confidence is low or just fallback to Gemini entirely? I’ll deliver a clear, comment-heavy n8n workflow JSON plus a README to get you running. I can demo typical queries yourself or live on request. Ready to start building this snappy, dual-AI support flow for you.

User Avatar
Flag of SERBIA

I can build the production-ready n8n workflow for RespondeYA!, connecting your chat webhook, identity verification, CRM/database retrieval, Claude, and Gemini into a fast, controlled support pipeline. The implementation will include: • Secure, minimal-field data retrieval from REST or SQL sources • Prompt templates that enforce response tone, formatting, and privacy boundaries • Confidence-based fallback and policy escalation between Claude and Gemini • Retries, timeouts, structured error handling, and review-safe logging • Configurable model ordering so switching primary and fallback models is straightforward • Environment-variable configuration, clear node naming, comments, and an import-ready README • A walkthrough demonstrating order, booking, and delivery-address scenarios I’ll design the flow around the under-five-second target while ensuring sensitive customer data is never unnecessarily exposed to logs or model context. I can also include validation and response safeguards so unsupported requests are escalated rather than guessed. The final handoff will be practical to deploy: workflow JSON, setup instructions, credential requirements, test cases, and a recorded or live demonstration.

内容来自 Freelancer News,版权归原作者所有。

阅读原文 ↗

把喜欢的内容,留在你的新标签页。

打开我的订阅